Steady State
← Back to Markov Chains
The long-run probability distribution that a Markov chain converges to regardless of the initial state (for ergodic chains). Found by solving pi * P = pi. The steady state represents the proportion of time the chain spends in each state over an infinite horizon.